Peritraumatic Distress among Chinese Canadians during the Early Lockdown Stage of the COVID-19 Pandemic: Sociodemographic and Pandemic-Related Predictors

Round 1

Reviewer 1 Report

Inconsistency among the sections, given by late descriptions of variables that were not included in the aims of the study( or introduced).

Limited information on the invitation/selection of participants, which could introduce bias. 

General conclusions, not informing the reader on what is the study adding to the literature. . 

Thank you for  the opportunity to review the document entitled “Peritraumatic Distress among Chinese Canadians during the Early Lockdown Stage of the COVID-19 Pandemic: Sociodemographic & Pandemic-related Predictors”, which was aimed to “1) identify sociodemographic predictors for psychological distress; and 2) determine COVID-19-related predictors (e.g., COVID-19-related experiences, perceptions, and safety behaviours) for the psychological distress of Chinese residents in Canada.”

The topic is interesting, and the results are relevant, however some concerns preclude its publication as presented.

-          The description of the aims of the study requires clarification. The variables selected as moderators (i.e. coping and satisfaction) are not included in the aims of the study nor described in the introduction. Similarly, the variable racial discrimination is missing in the aims of the study.

-          The cut-points to classify the age groups can be misleading and could interfere with comparisons with other studies, otherwise, an appropriate reference should be provided (according to the aims of the study instead of reproductive health). In addition, the intervals described in the text are not consistent with the intervals described in table 1.

-          The description of the invitation procedure and a flow chart to explain the full selection procedure are missing; both are essential to discuss potential bias.

-          The attrition procedure and the number of missing responses should be described.

-          Further discussion is required on the underrepresentation of social groups with factors that could imply proclivity for distress (e.g. low education, unemployment).

-          The conclusions should focus on the assessments /results of the study, to underline what the study added to the already existing literature.

-          The readers would appreciate further selection of the references provided for specific facts  in the introduction, according to the results and strength of each study.   

Author Response

Reviewer 1:

1. The description of the aims of the study requires clarification. The variables selected as moderators (i.e. coping and satisfaction) are not included in the aims of the study nor described in the introduction. Similarly, the variable racial discrimination is missing in the aims of the study.

Response: we have included the requested information in the introduction (section 1. 5 & 1.6)

2. The cut-points to classify the age groups can be misleading and could interfere with comparisons with other studies, otherwise, an appropriate reference should be provided (according to the aims of the study instead of reproductive health). In addition, the intervals described in the text are not consistent with the intervals described in table 1.

Response: We agree that the age cut-off, particularly for young group (i.e., 35), might not be consistent with most other studies, but please note that it is consistent with other publications (e.g., Yu et al., 2022). It is done this way mainly to ensure a sufficient group sample size for young adults (n = 43.9%) for any meaningful statistical comparisons with other age groups. We also fixed the inconsistent wording for the age group cut-off age in the text (section 2.3)

3. The description of the invitation procedure and a flow chart to explain the full selection procedure are missing; both are essential to discuss potential bias.

Response: To best protect the privacy and confidentiality, the recruitment was not invitation-based. We recruited participants openly and widely through local community email list, social media posts, websites, and internet. Participants anonymously participated through a link included in our recruitment post. We also added a flowchart (i.e., Figure 1) to illustrate the selection procedure (section 2.1)

4. The attrition procedure and the number of missing responses should be described.

Response: We added a flow chat (i.e., Figure 1) and some further clarifications in the “Participants” section to describe the nature of missing responses ( Section 2.1. Participants). Please note the survey participation is anonymous and voluntary. Participants can withdraw at any time point or skip any questions without disclosing the reason. Considering that the sociodemographic questions were completed as the last section of the survey. Those individuals excluded due to missing data points (n=199) barely completed any other sociodemographic variables, thus unfortunately it is impossible to assess the attrition effects based on sociodemographic variables.

5. Further discussion is required on the underrepresentation of social groups with factors that could imply proclivity for distress (e.g. low education, unemployment).

Response: we identified this as a limitation in the discussion section (section 4.1 & section 5).

6. The conclusions should focus on the assessments /results of the study, to underline what the study added to the already existing literature.

Response: we revised the conclusion section to stick to the results of the study, with a highlight on the novel contribution to the existing literature. (section 6).

7. The readers would appreciate further selection of the references provided for specific facts in the introduction, according to the results and strength of each study.   

Response: we enriched the introduction with more relevant references in light of the strength and results of the study. For example, we added a section on life satisfaction and coping (section 1.5).

Reviewer 2 Report

Author Response

Reviewer 2:

1. Is the research design appropriate and are the methodsadequately described? NO

in the methodological part it is not very clear how to respond to the various items. The study combines dichotomous questions with Likert scale responses. Why did the authors construct dichotomous questions that limit and condition the participants' responses? Thoughts, emotions and behaviors regarding such a complex phenomenon as Covid cannot be limited to a yes/no. in that period people found themselves trying different strategies, having conflicting thoughts and confused feelings. etc. In the psychological field, responses on liker scales are preferable, which then allow them to be correlated with the scores of other tests. How and on the basis of which theories or studies the items were constructed must be explained. The study remains confusing

Response: Thank you for the comment. Please note that all the continuous predictors (listed in Table 2) are based on responses to a 5-point or a 7-point Likert Scale. The SWLS and the CPID are standard measures based on Likert Scale responses. Most Dichotomous predictors are questions either from an overarching multiple option YES/NO selection question (e.g.,  whether each coping behaviour was engaged or not), or simplified from a multiple-option question to best capture the variance in CPDI and for presentation clarity consideration (e.g., sociodemographic variables such as age group). The related information and the data clearing procedure is described in the Materials section. In any event, we acknowledged that this might have restricted the variance of the results and identified this in the limitation section as you recommended (section 5).

2. Are the resultspresented clearly and in sufficient detail, are the conclusionssupported by the results and are they put into context within the existing literature? NO

Some analysis choices should be reviewed and correlational analyzes included which allow you to select which variables to include in the models. The results show few and small significances

Response: We did run ANOVAs (Table 1 & 4) to identify potential categorical predictors and correlations (Table 2) to identify potential continuous predictors to be entered in the regression models (Table 3 & 5). We appreciate your input and suggestion on the correlation analysis based on aggregated sum data and did give it further thoughts. However,  without clear theoretical justifications on such aggregation, we decided to stay the way it is. However, we would be happy to add any relevant further analyses that might enhance the statistical analysis of the current study.

3. Does this article provide a relevant contributionto the scientific discussion of this topic? NO

Compared to such a large sample, the significance of the results is low which implies: either that the tools have been constructed poorly or that the analysis choices are not the best

Response: We did conduct thorough preliminary analyses to streamline the data analysis approach. We also followed a convention in the field to identify potential predictors through ANOVA and correlation analysis to be entered in the linear regression models. The CPDI and SWLS are both standardised scales widely used in related literature. If there are any specific suggestions that you think might make the survey or analysis better address the proposed questions, we would appreciate your input and accommodate them in our future studies.

Major comments:

1. The study conducted by the authors appears interesting even though the data collected is from approximately 4 years ago. The idea of ​​considering coping strategies is very interesting. The studies by Vagni et al (2020; 2021), including those conducted on the general population (2022) highlighted how coping strategies but above all resilience skills had reduced levels of stress and secondary trauma.

Response: Thank you for pointing this out. We cited Vagni et al (2020; 2021) in the introduction and discussion to strengthen our study rationale (section 1.5).

2. In the methodological part it is not very clear how to respond to the various items. The study combines dichotomous questions with Likert scale responses. Why did the authors construct dichotomous questions that limit and condition the participants' responses? Thoughts, emotions and behaviors regarding such a complex phenomenon as Covid cannot be limited to a yes/no. in that period people found themselves trying different strategies, having conflicting thoughts and confused feelings. etc. In the psychological field, responses on liker scales are preferable, which then allow them to be correlated with the scores of other tests.

Response: see responses to the “Method” question above.

3. dichotomous answers condition and limit the analysis processes. in fact, the results show few and low significances The Anovas relating to the CPDI (table 1) are also few and of small size and significance. It is appropriate for the authors to report normality tests of the variables to verify the distributions.
   Given that the results show few and small significances, it appears necessary to first carry out a correlational analysis. By putting together dichotomous variables with scores obtained from the sum of several items, a Spearman analysis appears more appropriate. Based on these significances, build regression models. The regressions report little significance which in my opinion depends on the way in which the questions were constructed. The limits section should be expanded.

Response: see responses to the “results” question above.

4. The authors define coping strategies as actions that are not true coping strategies. They should also explain on what theories, studies or other factors they constructed these items. The coping strategies appear far too generic: is reading considered avoidant coping or problem solving? The system methodologically must absolutely be expanded. the results should also be compared with other studies that have used coping strategies in managing coping.

Response: Thank you for the correction. Following your comment, we rephrased “coping strategy” as “coping behaviour endorsement” to be more specific and correctly reflect what we measured in this study. We have corrected this throughout the document. 

Detailed comments: Explanations are needed on how the research items were constructed. It is necessary to report normality tests of the variables. insert correlational analyzes and review the statistical analysis choices on the basis of them  

Response: The outcome measure (CPDI) is a widely used standardised measure for distress. The moderator SWLS is also a standardised scale for life satisfaction. All the other items (sociodemographic and COVID-experience) were structured based on previous studies (Qian et al., 2020; Qiu et al., 2020). We added a sentence in the material section of the revised manuscript. Additionally, the outcome variable (CPDI) and moderator SWLS are both normally distributed, as indicated by the frequency analysis (CPDI: Skewness = 0.74; Kurtosis = 0.83; SWLS: Skewness = -0.51; Kurtosis = 0.16).

Reviewer 3 Report

Major Revision is needed 

Please see the attachment below

Comments for author File: Comments.pdf

Author Response

Reviewer 3:

Is the research design appropriate and are the methods adequately described? NO

1. How the authors ensure the participants are those reside in Canada not any countries outside the country?

Response: We have a screening question on whether participants identify them as Chinese living in Canada. We also asked them to identify their current resident district (province/area) in Canada.

2. The description of variables in paragraphs form are complicated, perhaps, you can presented them in a more systemtic way such as in table form and point form for their options, this facilitate the readers what were included in the analysis.

Response: All the variables described in the method section are also displayed in the Table with the corresponding options or scale information (Table 1-5). We would be happy to provide the survey in its original format if requested.

Are all of the cited references relevant to the research? No

- citation shown in main text do not tally with reference (format issue).

Response: we thoroughly checked all the citations and references to ensure they match. Thank you for bringing this to our attention

Round 2

Reviewer 1 Report

The main concerns were solved.

The manuscript was improved.

Author Response

Thank you for your comments, we changed the manuscript accordingly.

Reviewer 2 Report

Author Response

Thank you for your comments, we changed the manuscript accordingly.

Reviewer 3 Report

1. The paper utilized hierachical regression analyses however the methodological details specific to this aspect are absent. Despite my earlier reminder to include this information in the method section, my comment was purposely omitted from the rebuttal letter. This omission raises concerns about the reproducibility of the study's results. 

2. In my previous feedback to the author, i emphasized the importance of discussing the implications of the study result for psychological or community practices, unfortunately, these suggestions were not integrated into the revised manuscript and these comments were overlooked. Including this discussion is vital for providing valuable insights into the practice applications of the study results. 

Same as above. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf